Feeding Tube Clinic Effect on Nutrition.

BACKGROUND: Optimizing nutrition is essential for recovery after major surgery or severe illness. Feeding tubes (FT) can be placed in patients limited by oral enteral nutrition. Given the myriad of locations in which these procedures are performed (radiology, intensive care unit, and endoscopy suite), routine follow-up is challenging. The purpose of this study was to evaluate the impact of an FT clinic on nutrition. We hypothesized that enrollment in the FT clinic would result in improved nutritional outcomes. METHODS: Retrospective review of Veteran Affairs Medical Center patients with FTs placed from January 2010 to January 2020. Demographics and body mass index (BMI) were recorded. Serum albumin recorded within 1 month of tube placement was compared to within 1 month of tube removal, death, or at the end of the study period. FT clinic participation required at least 2 visits. Indications for FT placement and duration were recorded. Patients were excluded when both BMI and albumin values were incomplete, and if FTs were placed for decompression. RESULTS: Ninety-three patients underwent FT placement during the study period; 5 (5%) were excluded. The average age was 64.8±9.7 years, with the majority being male, 85 patients (97%). Eighteen (20%) patients were seen in the FT clinic (FTC) and 70 (80%) were managed outside of FTC (nFTC). There were no differences in age, gender, or indication for FT. Mean albumin increased 0.42±0.85 g/dL in the FTC group versus -0.07±0.72 g/dL in the nFTC group (P=0.037). The FTC group BMI increased, 0.38 kg/m2 vs. -1.48 kg/m2 in nFTC patients, P=0.041. The FTC patients maintained their tubes longer (36.5 vs. 7.0 mo, P=0.0014). CONCLUSIONS: Patients managed in a dedicated FT clinic experienced an improvement in their serum albumin values and increases in their BMI. In addition, they also maintained their FTs longer. To optimize nutrition and reduce weight loss, patients who require FTs should be enrolled in a dedicated FT clinic.

Are Functional Performance Test Scores Better When Compared to Baseline or Contralateral Limb Scores Following LE Injury in Adolescent Athletes?

Background: Functional performance tests (FPT) have been used with athletes following an injury to determine readiness to return-to-play (RTP), usually using limb symmetry indices to the contralateral limb or a baseline score. There is not a consensus as to which criterion scores are best compared. Hypothesis/Purpose: This study aimed to compare common functional performance test scores from injured athletes at the time of release to RTP to both preseason baseline scores and to the contralateral limb. It was hypothesized that using baseline scores for comparison would be more responsive to residual deficits following injury than using the contralateral limb. Study Design: Prospective longitudinal cohort study. Methods: High school athletes (n=395) from all varsity sports completed a battery of FPTs including the Y-Balance Test (YBT), single limb hop tests and T-Test for agility (TT) during their preseason to establish baseline data. Injured athletes (n=19) were re-tested using all FTP's again at the time of RTP. Paired t-tests were used to detect if significant (p\<0.05) residual deficits were present at time of RTP when compared to baseline and to contralateral scores on FPTs. Results: Differences in YBT scores were found in the anterior direction only (p=0.021) when comparing RTP to preseason, but there were no differences when compared to RTP data for the contralateral limb. Differences were detected with the single leg hop test (p = 0.001) when comparing the RTP to preseason and were also detected in both the single leg hop (p= 0.001) and triple hop (p=0.018) when compared to the contralateral limb. Differences in TT scores were detected when comparing RTP to preseason for cutting first with both the unaffected (p = 0.019) and affected (p = 0.014) limbs. Conclusions: The YBT in the anterior direction and the TT are better able to detect residual deficits when comparing RTP to preseason scores. Hop tests are better able to detect deficits when compared to the contralateral limb. These results could make preseason testing more efficient when creating a reference for determining RTP readiness following lower extremity injury.

FPGA-based fast bin-ratio spiking ensemble network for radioisotope identification.

In this work, we demonstrate the training, conversion, and implementation flow of an FPGA-based bin-ratio ensemble spiking neural network applied for radioisotope identification. The combination of techniques including learned step quantisation (LSQ) and pruning facilitated the implementation by compressing the network's parameters down to 30% yet retaining the accuracy of 97.04% with an accuracy loss of less than 1%. Meanwhile, the proposed ensemble network of 20 3-layer spiking neural networks (SNNs), which incorporates 1160 spiking neurons, only needs 334 µs for a single inference with the given clock frequency of 100 MHz. Under such optimisation, this FPGA implementation in an Artix-7 board consumes 157 µJ per inference by estimation.

NODAL variants are associated with a continuum of laterality defects from simple D-transposition of the great arteries to heterotaxy.

BACKGROUND: NODAL signaling plays a critical role in embryonic patterning and heart development in vertebrates. Genetic variants resulting in perturbations of the TGF-ß/NODAL signaling pathway have reproducibly been shown to cause laterality defects in humans. To further explore this association and improve genetic diagnosis, the study aims to identify and characterize a broader range of NODAL variants in a large number of individuals with laterality defects. METHODS: We re-analyzed a cohort of 321 proband-only exomes of individuals with clinically diagnosed laterality congenital heart disease (CHD) using family-based, rare variant genomic analyses. To this cohort we added 12 affected subjects with known NODAL variants and CHD from institutional research and clinical cohorts to investigate an allelic series. For those with candidate contributory variants, variant allele confirmation and segregation analysis were studied by Sanger sequencing in available family members. Array comparative genomic hybridization and droplet digital PCR were utilized for copy number variants (CNV) validation and characterization. We performed Human Phenotype Ontology (HPO)-based quantitative phenotypic analyses to dissect allele-specific phenotypic differences. RESULTS: Missense, nonsense, splice site, indels, and/or structural variants of NODAL were identified as potential causes of heterotaxy and other laterality defects in 33 CHD cases. We describe a recurrent complex indel variant for which the nucleic acid secondary structure predictions implicate secondary structure mutagenesis as a possible mechanism for formation. We identified two CNV deletion alleles spanning NODAL in two unrelated CHD cases. Furthermore, 17 CHD individuals were found (16/17 with known Hispanic ancestry) to have the c.778G > A:p.G260R NODAL missense variant which we propose reclassification from variant of uncertain significance (VUS) to likely pathogenic. Quantitative HPO-based analyses of the observed clinical phenotype for all cases with p.G260R variation, including heterozygous, homozygous, and compound heterozygous cases, reveal clustering of individuals with biallelic variation. This finding provides evidence for a genotypic-phenotypic correlation and an allele-specific gene dosage model. CONCLUSION: Our data further support a role for rare deleterious variants in NODAL as a cause for sporadic human laterality defects, expand the repertoire of observed anatomical complexity of potential cardiovascular anomalies, and implicate an allele specific gene dosage model.

An Efficient and Effective Image Decolorization Algorithm Based on Cumulative Distribution Function.

Image decolorization is an image pre-processing step which is widely used in image analysis, computer vision, and printing applications. The most commonly used methods give each color channel (e.g., the R component in RGB format, or the Y component of an image in CIE-XYZ format) a constant weight without considering image content. This approach is simple and fast, but it may cause significant information loss when images contain too many isoluminant colors. In this paper, we propose a new method which is not only efficient, but also can preserve a higher level of image contrast and detail than the traditional methods. It uses the information from the cumulative distribution function (CDF) of the information in each color channel to compute a weight for each pixel in each color channel. Then, these weights are used to combine the three color channels (red, green, and blue) to obtain the final grayscale value. The algorithm works in RGB color space directly without any color conversion. In order to evaluate the proposed algorithm objectively, two new metrics are also developed. Experimental results show that the proposed algorithm can run as efficiently as the traditional methods and obtain the best overall performance across four different metrics.

Evaluation of veteran community care outcomes after coronary artery bypass grafting: a retrospective pilot cohort.

For Veterans who cannot be seen in a timely fashion or must travel long distances to be seen, the Veterans Health Administration (VHA) offers funded care in the community. The use of this program has rapidly increased; however, there have been no systematic evaluations of surgery specific metrics such as perioperative complications, mortality and timeliness of care. To evaluate this in cardiac surgery patients, we compared veterans undergoing coronary artery bypass grafting in the community to those remaining within the VHA. We identified 78 patients during calendar year 2018 meeting inclusion criteria. 41 underwent surgery in the community versus 37 in the VHA. There were no significant differences in baseline demographics including age, sex, race, ethnicity, comorbidities and surgical risk scores. With regard to perioperative outcomes, veterans who underwent surgery within the VHA had lower infection rates (17% vs. 0%, p = 0.008) and 30-day emergency department utilization (22% vs. 5%, p = 0.04). A longer median postoperative inpatient stay was also seen within the VHA (8 days vs. 6 days, p < 0.001). These findings suggest that the VHA may better serve Veterans and prevent adverse events after CABG, at the expense of prolonged hospitalization. More study is needed to validate the findings of this pilot study.

Economic risks hidden in local water pollution and global markets: A retrospective analysis (1995-2010) and future perspectives on sustainable development goal 6.

Pollution from untreated wastewater discharges depletes clean water supply for humans and the environment. It poses adverse economic impacts by determining agricultural yields, manufacturing productivity, and ecosystem functionality. Current studies mainly focus on quantity-related water scarcity assessment. It is unknown how low water quality amplifies local water stress and induces cascading economic risks globally. In this study, we estimated both quality and quantity-related water scarcity index (WSI), local economic water scarcity risk (WSR), and cascading virtual WSR evident in global trade markets across 40 major economies from 1995 to 2010. We find developing countries, e.g., India and China, witnessed fast growth in both quantity and quality-related WSI. Major developed economies, e.g., the US and Germany, experienced a modest increase in water stress but alleviated quality-related risks. Local economic risk (WSR) grew from $116B to $380B, with quality-related risks rising from 20 % to 30 %. Virtual economic WSR in global supply chains increased from $39B to $160B, with quality-related risks increasing from 19 % to 27 %. China became the top exporter of economic WSR, ranked above the US, France, and Japan, and the second-largest position as an importer, trailing only the US. We finally conducted scenario modeling by 2030, assuming different progresses on SDG 6 targets. The findings suggest that only the most ambitious progress in both water quality enhancement and efficiency improvement helps to alleviate ∼20 % economic WSR globally. Our findings underscore the necessity for strategies that integrate management of untreated wastewater flows, improved water use efficiency, and diversification of supply chain networks to enhance global economic resilience to water challenges in the future.

Surgical Fires Involving Alcohol-Based Preparation Solution, 1991-2020.

BACKGROUND: Alcohol-based skin preparations were first approved for surgical use in 1998 and have since become standard in most surgical fields. The purpose of this report is to examine incidence of surgical fires because of alcohol-based skin preparation and to understand how approval and regulation of alcohol-based skin preparations impacted trends in fires over time. METHODS: We identified all reported surgical fires resulting in patient or staff harm from 1991 through 2020 reported to the Food and Drug Administration's Manufacturer and User Facility Device Experience (MAUDE) database. We examined incidence of fires because of these preparations, trends after approval and regulation, and common causes. RESULTS: We identified 674 reports of surgical fires resulting in harm to patients and surgical personnel, in which 84 involved an alcohol-based preparation. The time-adjusted model shows that from 1996 through 2006, there was a 26.4% increase in fires followed by a 9.7% decrease from 2007 to 2020. The decrease in fires was most rapid for head and neck and upper aerodigestive tract surgeries. Qualitative content analysis revealed improper surgical site preparation as well as close proximity of surgical sites to an oxygen source as the most common causes of fires. CONCLUSION: Since FDA approval, alcohol-based preparation solutions have been associated with a significant percentage of surgical fires. Warning label updates from 2006 to 2012 coupled with increased awareness efforts of associated risks of alcohol-based surgical solutions likely contributed to the decrease in fires. Improper surgical site preparation technique and close proximity of surgical sites to oxygen continue to be risk factors for fires. LEVEL OF EVIDENCE: 4 Laryngoscope, 134:607-613, 2024.

Hypoxia following warden procedure: evaluation and percutaneous treatment.

Partial anomalous venous connection with sinus venosus atrial septal defect is repaired with different approaches including the Warden procedure. Complications include stenosis of the superior caval vein and pulmonary venous baffle; however, cyanosis is rarely seen post-operatively. We report a patient presenting with cyanosis 5 years after a Warden, which was treated with a transcatheter approach.

The Laparoscopy in Biliary Exploration Research and Training Initiative (LIBERTI) trial: simulator-based training for laparoscopic management of choledocholithiasis.

BACKGROUND: Laparoscopic cholecystectomy with common bile duct exploration (LCBDE) is equivalent in safety and efficacy to endoscopic retrograde cholangiopancreatography (ERCP) plus laparoscopic cholecystectomy (LC) while decreasing number of procedures and length of stay (LOS). Despite these advantages LCBDE is infrequently utilized. We hypothesized that formal, simulation-based training in LCBDE would result in increased utilization and improve patient outcomes across participating institutions. METHODS: Data was obtained from an on-going multi-center study in which simulator-based transcystic LCBDE training curricula were instituted for attending surgeons and residents. A 2-year retrospective review of LCBDE utilization prior to LCBDE training was compared to utilization up to 2 years after initiation of training. Patient outcomes were analyzed between LCBDE strategy and ERCP strategy groups using χ2, t tests, and Wilcoxon rank tests. RESULTS: A total of 50 attendings and 70 residents trained in LCBDE since November 2020. Initial LCBDE utilization rate ranged from 0.74 to 4.5%, and increased among all institutions after training, ranging from 9.3 to 41.4% of cases. There were 393 choledocholithiasis patients analyzed using LCBDE (N = 129) and ERCP (N = 264) strategies. The LCBDE group had shorter median LOS (3 days vs. 4 days, p < 0.0001). No significant differences in readmission rates between LCBDE and ERCP groups (4.7% vs. 7.2%, p = 0.33), or in post-procedure pancreatitis (0.8% v 0.8%, p > 0.98). In comparison to LCBDE, the ERCP group had higher rates of bile duct injury (0% v 3.8%, p = 0.034) and fluid collections requiring intervention (0.8% v 6.8%, p < 0.009) secondary to cholecystectomy complications. Laparoscopic antegrade balloon sphincteroplasty had the highest technical success rate (87%), followed by choledochoscopic techniques (64%). CONCLUSION: Simulator-based training in LCBDE results in higher utilization rates, shorter LOS, and comparable safety to ERCP plus cholecystectomy. Therefore, implementation of LCBDE training is strongly recommended to optimize healthcare utilization and management of patients with choledocholithiasis.

Outpatient surgery postoperative ambulation and emergency department utilization.

BACKGROUND: The ability to ambulate is an important indicator for wellness and quality of life. A major health event, such as a surgery, can derail this ability, and return to preoperative walking ability is a marker for recovery. Self-reported walking measurements by patients are subject to bias, thus wearable technology such as activity monitors have risen in popularity. We evaluated postoperative ambulation using an accelerometer in outpatient general surgery procedures with the hypothesis that those patients with less postoperative ambulation were at risk for adverse outcomes. METHODS: A retrospective review of patients undergoing outpatient abdominal surgeries from November 2016 to July 2019 at a Veteran Affairs Medical Center. Patients wore an accelerometer preoperatively and postoperatively to measure their ambulation (steps/day). Outcome measures were 30-day readmissions and Emergency Department (ED) utilization. Postoperative ambulation was defined as daily percentages of their preoperative baseline. Patients without preoperative baseline data, > 3 missing days or any missing days prior to reaching baseline were excluded. RESULTS: One-hundred-six patients underwent outpatient abdominal surgery. Twenty-two patients were excluded. Patients stratified into adult (18-64 years, 44 patients, 52%) and geriatric (≥ 65 years, 40 patients, 48%) cohorts. Geriatric patients were less likely to meet their preoperative baseline by postoperative day 7, 35% vs 61%, p = 0.016. Adult patients who failed to meet their preoperative baseline in first postoperative week had higher ED utilization; 4 (24%) vs 1 (4%), p = 0.04. Geriatric patients who failed to meet their baseline trended toward increased ED utilization; 5 (19%) vs. 1 (7%), p = 0.31. CONCLUSION: Patients aged < 65 who fail to return to their preoperative daily step count within one week of outpatient abdominal surgery are 6× more likely to be seen in the ED. Postoperative ambulation may be able to predict ED utilization and recovery after outpatient surgery.

Telehealth for general surgery postoperative care.

BACKGROUND: Telehealth utilization rapidly increased following the pandemic. However, it is not widely used in the Veteran surgical population. We sought to evaluate postoperative telehealth in patients undergoing general surgery. METHODS: Retrospective review of Veterans undergoing general surgery at a level 1A VA Medical Center from June 2019 to September 2021. Exclusions were concomitant procedure(s), discharge with drains or non-absorbable sutures/staples, complication prior to discharge or pathology positive for malignancy. RESULTS: 1075 patients underwent qualifying procedures, 124 (12 â€‹%) were excluded and 162 (17 â€‹%) did not have follow-up. 443 (56 â€‹%) patients followed-up in-person (56 â€‹%) vs 346 (44 â€‹%) via telehealth. Telehealth patients had a lower rate of complications, 6 â€‹% vs 12 â€‹%, p â€‹= â€‹0.013. There were no significant differences in ED visits, 30-day readmission, postoperative procedures or missed adverse events. CONCLUSION: Telehealth follow-up after general surgical procedures is safe and effective. Postoperative telehealth care should be considered after low-risk general surgery procedures.

Impact of ISP Tuning on Object Detection.

In advanced driver assistance systems (ADAS) or autonomous vehicle research, acquiring semantic information about the surrounding environment generally relies heavily on camera-based object detection. Image signal processors (ISPs) in cameras are generally tuned for human perception. In most cases, ISP parameters are selected subjectively and the resulting image differs depending on the individual who tuned it. While the installation of cameras on cars started as a means of providing a view of the vehicle's environment to the driver, cameras are increasingly becoming part of safety-critical object detection systems for ADAS. Deep learning-based object detection has become prominent, but the effect of varying the ISP parameters has an unknown performance impact. In this study, we analyze the performance of 14 popular object detection models in the context of changes in the ISP parameters. We consider eight ISP blocks: demosaicing, gamma, denoising, edge enhancement, local tone mapping, saturation, contrast, and hue angle. We investigate two raw datasets, PASCALRAW and a custom raw dataset collected from an advanced driver assistance system (ADAS) perspective. We found that varying from a default ISP degrades the object detection performance and that the models differ in sensitivity to varying ISP parameters. Finally, we propose a novel methodology that increases object detection model robustness via ISP variation data augmentation.

Experimental Investigation of Acoustic Features to Optimize Intelligibility in Cochlear Implants.

Although cochlear implants work well for people with hearing impairment in quiet conditions, it is well-known that they are not as effective in noisy environments. Noise reduction algorithms based on machine learning allied with appropriate speech features can be used to address this problem. The purpose of this study is to investigate the importance of acoustic features in such algorithms. Acoustic features are extracted from speech and noise mixtures and used in conjunction with the ideal binary mask to train a deep neural network to estimate masks for speech synthesis to produce enhanced speech. The intelligibility of this speech is objectively measured using metrics such as Short-time Objective Intelligibility (STOI), Hit Rate minus False Alarm Rate (HIT-FA) and Normalized Covariance Measure (NCM) for both simulated normal-hearing and hearing-impaired scenarios. A wide range of existing features is experimentally evaluated, including features that have not been traditionally applied in this application. The results demonstrate that frequency domain features perform best. In particular, Gammatone features performed best for normal hearing over a range of signal-to-noise ratios and noise types (STOI = 0.7826). Mel spectrogram features exhibited the best overall performance for hearing impairment (NCM = 0.7314). There is a stronger correlation between STOI and NCM than HIT-FA and NCM, suggesting that the former is a better predictor of intelligibility for hearing-impaired listeners. The results of this study may be useful in the design of adaptive intelligibility enhancement systems for cochlear implants based on both the noise level and the nature of the noise (stationary or non-stationary).

Stray energy injury during robotic versus laparoscopic inguinal hernia repair: a randomized controlled trial.

BACKGROUND: Stray energy transfer from monopolar instruments during laparoscopic surgery is a recognized cause of potentially catastrophic complications. There are limited data on stray energy injuries in robotic surgery. We sought to characterize stray energy injury in the form of superficial burns to the skin surrounding laparoscopic and robotic trocar sites. Our hypothesis was that stray energy burns will occur at all laparoscopic and robotic port sites. METHODS: We conducted a prospective, randomized controlled trial of patients undergoing elective unilateral inguinal hernia repair at a VAMC over a 4-year period. Surgery was performed via transabdominal preperitoneal approach either laparoscopic-assisted (TAPP) or robotic-assisted (rTAPP). A monopolar scissor was used to deliver energy at 30W coagulation for all cases. At completion of the procedure, skin biopsies were taken from all the port sites. A picro-Sirius red stain was utilized to identify thermal injury by a blinded pathologist. RESULTS: Over half (54%, 59/108) of all samples demonstrated thermal injury to the skin. In the laparoscopic group, 49% (25/51) samples showed thermal injury vs. 60% (34/57) in the robotic group (p = 0.548). The camera port was the most frequently involved with 68% (13/19) rTAPP samples showing injury vs. 47% (8/17) in the TAPP group (p = 0.503). There was no difference in the rate of injury at the working port site (rTAPP 53%, 10/19 vs. TAPP 47%, 8/17; p = 0.991) or the assistant port site (rTAPP 58%, 11/19 vs. TAPP 53%, 9/17; p = 0.873). CONCLUSIONS: Stray energy causes thermal injury to the skin at port sites in the majority robotic laparoscopic TAPP inguinal hernia repairs. There is no difference in stray energy transfer between the laparoscopic and robotic platform. This is the first study to confirm in-vivo transfer of stray energy during robotic surgical procedures. More study is needed to determine the clinical significance of these thermal injuries.

PRDM16 Deletion Is Associated With Sex-dependent Cardiomyopathy and Cardiac Mortality: A Translational, Multi-Institutional Cohort Study.

BACKGROUND: 1p36 deletion syndrome can predispose to pediatric-onset cardiomyopathy. Deletion breakpoints are variable and may delete the transcription factor PRDM16. Early studies suggest that deletion of PRDM16 may underlie cardiomyopathy in patients with 1p36 deletion; however, the prognostic impact of PRDM16 loss is unknown. METHODS: This retrospective cohort included subjects with 1p36 deletion syndrome from 4 hospitals. Prevalence of cardiomyopathy and freedom from death, cardiac transplantation, or ventricular assist device were analyzed. A systematic review cohort was derived for further analysis. A cardiac-specific Prdm16 knockout mouse (Prdm16 conditional knockout) was generated. Echocardiography was performed at 4 and 6 to 7 months. Histology staining and qPCR were performed at 7 months to assess fibrosis. RESULTS: The retrospective cohort included 71 patients. Among individuals with PRDM16 deleted, 34.5% developed cardiomyopathy versus 7.7% of individuals with PRDM16 not deleted (P=0.1). In the combined retrospective and systematic review cohort (n=134), PRDM16 deletion-associated cardiomyopathy risk was recapitulated and significant (29.1% versus 10.8%, P=0.03). PRDM16 deletion was associated with increased risk of death, cardiac transplant, or ventricular assist device (P=0.04). Among those PRDM16 deleted, 34.5% of females developed cardiomyopathy versus 16.7% of their male counterparts (P=0.2). We find sex-specific differences in the incidence and the severity of contractile dysfunction and fibrosis in female Prdm16 conditional knockout mice. Further, female Prdm16 conditional knockout mice demonstrate significantly elevated risk of mortality (P=0.0003). CONCLUSIONS: PRDM16 deletion is associated with a significantly increased risk of cardiomyopathy and cardiac mortality. Prdm16 conditional knockout mice develop cardiomyopathy in a sex-biased way. Patients with PRDM16 deletion should be assessed for cardiac disease.

Losing weight to achieve joint or hernia surgery: is the intragastric balloon the answer?

BACKGROUND: Obesity is an epidemic, with its accompanying medical conditions putting patients at increased risk of postoperative complications. For patients undergoing elective surgery, preoperative weight loss provides an opportunity to decrease complications. We sought to evaluate the safety and efficacy of an intragastric balloon in achieving a body mass index (BMI) < 35 kg/m2 prior to elective joint replacement or hernia repair. METHODS: Retrospective review of all patients who had intragastric balloon placement at a level 1A VA medical center from 1/2019 to 1/2023. Patients who had a scheduled qualifying procedure (knee/hip replacement or hernia repair) and had a BMI > 35 kg/m2 were offered intragastric balloon placement to achieve 30-50lbs (13-28 kg) weight loss prior to surgery. Participation in a standardized weight loss program for 12 months was required. Balloons were removed 6 months after placement, preferentially concomitant with the qualifying procedure. Baseline demographics, duration of balloon therapy, weight loss and progression to qualifying procedure were recorded. RESULTS: Twenty patients completed intragastric balloon therapy and had balloon removal. Mean age 54 (34-71 years), majority (95%) male. Mean balloon duration was 200 ± 37 days. Mean weight loss was 30.8 ± 17.7lbs (14.0 ± 8.0 kg) with an average BMI reduction of 4.4 ± 2.9. Seventeen (85%) patients were successful, 15 (75%) underwent elective surgery and 2 (10%) were no longer symptomatic after weight loss. Three patients (15%) did not lose sufficient weight to qualify or were too ill to undergo surgery. Nausea was the most frequent side effect. One (5%) patient was readmitted within 30 days for pneumonia. DISCUSSION: Intragastric balloon placement resulted in an average 30lbs (14 kg) weight loss over 6 months allowing more than 75% of patients to undergo joint replacement or hernia repair at an optimal weight. Intragastric balloons should be considered in patients requiring 30-50lbs (13-28 kg) weight loss prior to elective surgery. More study is needed to determine the long-term benefit of preoperative weight loss prior to elective surgery.

BitBrain and Sparse Binary Coincidence (SBC) memories: Fast, robust learning and inference for neuromorphic architectures.

We present an innovative working mechanism (the SBC memory) and surrounding infrastructure (BitBrain) based upon a novel synthesis of ideas from sparse coding, computational neuroscience and information theory that enables fast and adaptive learning and accurate, robust inference. The mechanism is designed to be implemented efficiently on current and future neuromorphic devices as well as on more conventional CPU and memory architectures. An example implementation on the SpiNNaker neuromorphic platform has been developed and initial results are presented. The SBC memory stores coincidences between features detected in class examples in a training set, and infers the class of a previously unseen test example by identifying the class with which it shares the highest number of feature coincidences. A number of SBC memories may be combined in a BitBrain to increase the diversity of the contributing feature coincidences. The resulting inference mechanism is shown to have excellent classification performance on benchmarks such as MNIST and EMNIST, achieving classification accuracy with single-pass learning approaching that of state-of-the-art deep networks with much larger tuneable parameter spaces and much higher training costs. It can also be made very robust to noise. BitBrain is designed to be very efficient in training and inference on both conventional and neuromorphic architectures. It provides a unique combination of single-pass, single-shot and continuous supervised learning; following a very simple unsupervised phase. Accurate classification inference that is very robust against imperfect inputs has been demonstrated. These contributions make it uniquely well-suited for edge and IoT applications.

Pedestrian Crossing Intention Forecasting at Unsignalized Intersections Using Naturalistic Trajectories.

Interacting with other roads users is a challenge for an autonomous vehicle, particularly in urban areas. Existing vehicle systems behave in a reactive manner, warning the driver or applying the brakes when the pedestrian is already in front of the vehicle. The ability to anticipate a pedestrian's crossing intention ahead of time will result in safer roads and smoother vehicle maneuvers. The problem of crossing intent forecasting at intersections is formulated in this paper as a classification task. A model that predicts pedestrian crossing behaviour at different locations around an urban intersection is proposed. The model not only provides a classification label (e.g., crossing, not-crossing), but a quantitative confidence level (i.e., probability). The training and evaluation are carried out using naturalistic trajectories provided by a publicly available dataset recorded from a drone. Results show that the model is able to predict crossing intention within a 3-s time window.